Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Cadmium (Cd²⁺) is an environmentally widely dispersed, highly toxic heavy metal which, despite recent legislative measures to reduce human exposure, continues to represent a serious threat to human health (1). Several reports have suggested that exposure to Cd²⁺ dusts or fumes at the workplace can lead to the development of emphysema (2, 3). Cd²⁺ is also an important contaminant of tobacco and has been proposed to play a role in the pathogenesis of emphysema in cigarette smokers (4). In experimental animals, both emphysema (4) and fibrosis with (5) and without (6) airspace enlargement have been reported.

Emphysema is associated with permanent abnormal enlargement of the respiratory airspaces, accompanied by destruction of connective tissue proteins within the alveolar wall. The predominant hypothesis for the pathogenesis of this disease proposes that progressive destruction of the interstitium is due to an excess of proteinases compared with the availability of antiproteinases (7). A series of contradictory studies on the role of this mechanism in cadmium-induced lung disease has led to a reevaluation of this hypothesis (8).

Connective tissue proteins in the lung are not inert proteins but form highly dynamic structures which are continuously being synthesized and degraded throughout life. Imbalances due to changes in either synthesis, degradation, or both, may therefore lead to either net deposition or loss of connective tissue, and such changes are now generally felt to be important in the pathogenesis of pulmonary fibrosis and emphysema, respectively (11). Part of the action of Cd²⁺ in the lung may therefore be occurring via direct effects on fibroblasts, the main connective tissue producing cells in the lung. In support of this hypothesis, we have previously shown that Cd²⁺ inhibits fibroblast proliferation and procollagen synthesis (12). However, little is known about the effects of Cd²⁺ on the production of proteoglycans. As well as influencing tissue compliance, elasticity, and fluid balance as a result of their large hydrodynamic volumes, these molecules play a pivotal role in connective tissue repair by virtue of their ability to regulate cell proliferation, migration, and adhesion (13 and references therein). Furthermore, decorin, the major small proteoglycan present in fibrous connective tissue in close association with fibrillar collagen (14, 15), plays a crucial role in the organization of the newly formed connective tissue by influencing collagen fibrillogenesis and architecture (16).

Fibroblasts have been reported to produce a number of different proteoglycans in vitro, some of which are secreted into the culture medium whereas others remain associated with the membrane or the surrounding matrix (17). In this paper we show that Cd²⁺ is a potent inhibitor of fibroblast proteoglycan synthesis and that the effects of Cd²⁺ on procollagen and proteoglycan synthesis are mediated, at least in part, via reduced steady-state messenger RNA (mRNA) levels. These data support the hypothesis that Cd²⁺ plays an important role in the pathogenesis of emphysema in humans exposed to Cd²⁺ in cigarettes or at the workplace by inhibiting the production of connective tissue proteins.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials

General. Human fetal lung fibroblasts (HFL1) were obtained from the American Type Culture Collection (Rockville, MD). Sulfate-poor Dulbecco's modified Eagle's medium (DMEM) was purchased from Nordvac AB (Stockholm, Sweden). Newborn calf serum (NCS) was from Imperial Laboratories (Andover, UK). Cadmium chloride (CdCl₂) and urea were obtained from Sigma Chemical Company (Poole, Dorset, UK). L-[4-³H]-phenylalanine, carrier-free [³⁵S]-sulfate, and [³H]-glucosamine were from Amersham International (Aylesbury, UK). Diethylaminoethyl (DEAE)-cellulose was from Whatman (Maidstone, Kent, UK), and Biogel P6 was from BioRad (Solna, Sweden). Sephacryl S-500HR, Mono Q HR, Superose 6, Octyl Sepharose CL-4B, and Sephadex G-50 were from Pharmacia (Uppsala, Sweden). CHAPS was from Fluka Chemie AG (Buchs, Switzerland). Mulgophene (tridecyloxypolyethyleneoxy-ethanol) was obtained from GFA Corporation (Johanneshov, Sweden). Chondroitinase ABC (EC 4.2.2.4) and chondroitinase AC (EC 4.2.2.5) were from Seikakagu Kogyo Co. (Tokyo, Japan). Alcian blue (8GS; 1 A 288) was a product of Chroma-Gesellschaft (Köngen, Germany). Stock solutions of 8 M guanidinium chloride from Fluka Chemie AG (practical grade) were treated with activated charcoal before use. Stock solutions of 8 M urea were purified by ion-exchange resin treatment.

Complementary DNA (cDNA) probes. Probes used included a cDNA probe for biglycan representing a 603-bp XbaI-StuI fragment (18), a cDNA probe for decorin representing an 1,162-bp EcoRI insert (19); a 1,300-bp fragment of versican cDNA (Telios Pharmaceuticals, Inc., San Diego, CA); and a 900-bp PstI fragment of cDNA for glyceraldehyde phosphate dehydrogenase (GAPDH). Human pro-1(I) collagen cDNA, the insert of clone pHCAL 1, was digested with PvuII and PstI to release a 372-bp probe fragment (20).

Cell Culture Conditions

HFL1 fibroblasts were maintained in DMEM supplemented with 10% NCS (vol/vol). For protein synthesis experiments, cells were grown to confluence in 2.4-cm-diameter wells in DMEM supplemented with 5% NCS. Upon reaching visual confluence, the media were replaced with sulfate-poor DMEM (containing 0.11 mM sulfate) supplemented with 50 µg/ml ascorbic acid, 0.2 mM proline, and 2%(vol/vol) NCS. After a further 24 h, the media were replaced with sulfate-poor DMEM-2% NCS with and without CdCl₂. For the determination of total protein and proteoglycan synthesis, wells were supplemented with either 5 µCi L-[4-³H]-phenylalanine/ml or 200 µCi/ml [³⁵S]-sulfate. In certain instances, 50 µCi/ml [³H]-glucosamine was added for assessing proteoglycan sulfation.

Estimation of CdCl₂ Cytotoxicity

CdCl₂ cytotoxicity was assessed on fibroblasts grown to confluence and exposed to CdCl₂ for 24 h by measuring the release of cytoplasmic lactate dehydrogenase (LDH) and uptake of the supravital dye neutral red, according to previously described spectrophotometric methods (21, 22).

Determination of Procollagen Production

Procollagen production was assessed by measuring hydroxyproline in intact proteins as previously described (12). Briefly, cells were grown to confluence and incubated as described above. At the end of the incubation period, the cell layer and media were harvested together. Proteins were precipitated by addition of ethanol to a final concentration of 67% (vol/vol) at 4°C overnight and separated from free amino acids by filtration. Filters with adherent proteins were hydrolyzed in hydrochloric acid (6 M) at 110°C overnight. Hydrolysates were mixed with charcoal, filtered, and derivatized with 7-chloro-4-nitrobenzo-2-oxa-1,3-diazole prior to reverse-phase high-performance liquid chromatography for the isolation of hydroxyproline with an acetonitrile gradient as previously described (12).

Determination of Proteoglycan Metabolism

Culture medium. Cells were grown to confluence and pre-incubated as described above. At the end of the incubation period, the culture medium was harvested separately and diisopropylphosphorofluoridate was added at a final concentration of 1 mM. Proteoglycans from the medium were recovered by passage over columns (0.7 × 4 cm) of DEAE-cellulose (DE-52), which had been equilibrated in 6 M urea, 50 mM sodium acetate (pH 8.5), 5 mM N-ethyl maleimide, and 5 µg/ml ovalbumin. The columns were washed with 60 bed volumes of the same buffer followed by six bed volumes of the same solvent containing 0.4 M sodium acetate. Proteoglycans were eluted with 6 bed volumes of 4 M guanidinium chloride, 50 mM sodium acetate (pH 5.8), and 5 µg/ml ovalbumin as previously described (15).

Cell layer. The cell layer was washed with phosphate-buffered saline and extracted with 200 µl of a solution of ice-cold 4 M guanidinium chloride, 50 mM sodium acetate buffer (pH 5.8), 20 mM ethylenediamenetetraacetic acid (EDTA), 1% Triton X, 10 mM N-ethylmaleimide, and 5 µg/ml ovalbumin at 4°C overnight. The culture well was washed with a further 200-µl extraction solution, which was combined with the first cell extract. Cell extracts were diluted with 20 volumes of 6 M urea, 50 mM sodium acetate (pH 8.5), 5 mM N-ethylmaleimide, and 5 µg/ml ovalbumin supplemented with 0.1% Triton-X, and proteoglycans were eluted as described above.

Proteoglycan Separation

Gel chromatography on Sephacryl S-500 HR. Proteoglycans were further separated into large and small proteoglycans by gel chromatography on a column of Sephacryl S-500 HR with 4 M guanidinium chloride, 50 mM sodium acetate (pH 5.8), 0.1% Triton X-100, and 5 µg/ml ovalbumin, at a rate of 13 ml/h.

Hydrophobic interaction chromatography. Fractions with small proteoglycans from Sephacryl S-500 were further subjected to hydrophobic interaction chromatography on Octyl Sepharose CL-4B (0.5 × 10 cm Omnifit column) using an FPLC system (23). The column was equilibrated with 2 M guanidinium chloride, and small proteoglycans were eluted and collected in 0.5-ml fractions with a linear gradient of 2 to 6 M guanidinium chloride at a flow rate of 0.1 ml/min.

Identification of chondroitin/dermatan sulfate proteoglycans. Chondroitin/dermatan sulfate proteoglycans were identified by chondroitinase AC and/or chondroitinase ABC (5 U/ml) digestion performed at 37°C for 4 h in 10 mM Tris acetate and 10 mM NaF buffer, pH 7.3.

Determination of 4- and 6-sulfation in Chondroitin/Dermatan Sulfate

Ion-exchange FPLC on Mono Q HR. After ion-exchange chromatography on DE-52 (see above), hyaluronan was removed by ion-exchange chromatography on a Mono Q HR 5/5 column connected to an FPLC system. Samples were diluted 20-fold with 6 M urea, 50 mM sodium acetate (pH 5.8), and 0.1 M NaCl, and eluted using a NaCl gradient ranging from 0.1 to 1.5 M. Appropriate fractions were pooled and material was recovered by passage through a DE-52 column as described above.

Lichrosorb-NH₂. The 4-sulfated and 6-sulfated disaccharides from glycosaminoglycan chains were quantitated following digestion with chondroitinase AC/ABC (see above) and separation on a Lichrosorb-NH₂ column (4 mm × 250 mm; Merck connected to an FPLC system; LKB, Stockholm, Sweden) (24). After digestion, four volumes of ethanol were added for 4 h. Digests were centrifuged, supernatants were evaporated to dryness, and residues were dissolved in 100 mM sodium acetate, pH 5.0.

Analytical Methods

Alcian blue precipitation and agarose gel precipitation. Large proteoglycans were identified by precipitation with Alcian blue and further analyzed before and after digestion by agarose (2%) gel electrophoresis using a discontinuous buffer system with 10 to 100 mM Tris-acetate, pH 7.3, as previously described (25).

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Small proteoglycans were identified by SDS-PAGE as previously described (26). Samples were dissolved in 60 µl of a solution of 5% (wt/vol) SDS/20% (vol/vol) glycerol/4 mM EDTA/0.04% bromophenol blue/ 125 mM Tris (pH 5.8) and -mercaptoethanol at a final concentration of 10% (vol/vol).

Measurement of radioactivity. Radioactivity associated with an aliquot of the column eluent corresponding to the proteoglycan fraction or with sample hydrolysates of ethanol-precipitated proteins from cell cultures incubated with L-[4-³H]-phenylalanine (for total protein synthesis measurement) was quantified on a liquid scintillation counter. Following electrophoresis, the radioactivity associated with isolated proteoglycans was visualized and quantitated by a Fuji bio-imaging analyzer system (Fuji Photo Film Co., Ltd., Tokyo, Japan).

Determination of the DNA Content of the Cell Monolayer

All protein synthesis measurements are expressed per microgram of DNA in the cell layer as an estimate of cell number assessed in parallel cell cultures, using a spectophotometric assay as previously described (12).

Isolation and Analysis of RNA

Total RNA was isolated from cell cultures by guanidinium isothiocyanate and phenol/chloroform/isoamyl alcohol extraction as previously described (27). Approximately 10 µg of RNA per samples was electrophoresed on a 1% agarose-formaldehyde gel and transferred to nylon (Hybond^TM-N, Amersham International) filters. Filters were hybridized with [³²P]-labeled cDNA probes overnight in 50% formamide at 42°C and sequentially washed with 2× standard saline citrate (SSC)/0.1% SDS at 42°C followed with 0.2× SSC/0.1% SDS at 50°C.

Statistical Analysis

All data are presented as mean ± standard errors of the mean from six replicate cultures, unless otherwise indicated. Statistical evaluation was performed using an unpaired t test for single-group comparisons and Newman- Keuls one-way analysis of variance for multiple-group comparisons. The mean values of various parameters were said to be significantly different when the probability of the differences of that magnitude, assuming the null hypothesis to be correct, fell below 5% (i.e., P < 0.05).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

CdCl₂ Cytotoxicity

The effect of increasing concentrations of CdCl₂ on the release of cytoplasmic LDH and the uptake of the supravital dye, neutral red, after 24-h exposure to CdCl₂ by the cells used in this study is shown in Table 1. There were no significant effects on LDH release by these cells until 150 µM CdCl₂, whereas uptake of neutral red was significantly reduced from 100 µM CdCl₂ onward, with values reduced by 12 ± 3% (P < 0.05) relative to media controls.

Effect of CdCl₂ on Proteoglycan Synthesis and Total Protein Synthesis

Figure 1 compares the effect of CdCl₂ on total proteoglycan synthesis for proteoglycans associated with the media (panel A) and cell layer (panel B), and total protein synthesis (panel C). Cadmium had no effect on total proteoglycan synthesis in cells exposed to 10 µM CdCl₂. However, in cells exposed to 30 µM CdCl₂, proteoglycans secreted into the culture media were reduced by 36 µM CdCl₂, whereas proteoglycans associated with the cell layer were reduced by 42 ± 6% (both P < 0.01). In the same experiment, CdCl₂ also inhibited procollagen production as previously reported (12), with values reduced by 22 ± 6% (P < 0.05) and 69 ± 6% (P < 0.01) at 10 µM and 30 µM CdCl₂, respectively (data not shown). In contrast, total protein synthesis was unaffected at both doses of CdCl₂ examined.

View larger version (14K): [in this window] [in a new window]   Figure 1.   Effect of Cd2+ on total proteoglycan production associated with the culture medium (A) and with the cell layer (B) and total protein synthesis (C). Figure shows the effect of CdCl2 on total proteoglycan production for proteoglycans associated with the media (panel A) and the cell layer (panel B), as well as total protein synthesis. Proteoglycan synthesis is based on the incorporation of [35S]-sulfate into proteoglycan GAG chains following purification by ion-exchange chromatography on a DE-52 column, whereas total protein synthesis is based on the incorporation of [3H]-phenylalanine into ethanol-insoluble protein. All values are expressed per microgram of DNA measured in parallel cell cultures.

Effect of CdCl₂ on the Production of Large Proteoglycans

To determine which proteoglycans were affected by CdCl₂ treatment, proteoglycans recovered from the medium (Figure 2) and the cell layer (Figure 3) from media control cells and cells exposed to 30 µM CdCl₂ were separated according to size by gel chromatography. Large proteoglycans from the medium and the cell layer, as well as intermediate-sized proteoglycans, were further identified by electrophoresis after Alcian blue precipitation before and after chondroitinase ABC digestion (Figures 2 and 3, insets). These characterizations revealed that the major secreted proteoglycan was a large chondroitin/dermatan sulfate proteoglycan that accounted for 77% of proteoglycans eluting close to the void volume (Figure 2, component I). The remaining proteoglycans in this fraction were heparan sulfate proteoglycans with a predominant core protein of 250 kD (15).

View larger version (21K): [in this window] [in a new window]   Figure 2.   Gel chromatography on Sephacryl S-500 and electrophoresis of radioactively labeled proteoglycans from the culture medium from control and Cd2+-treated cultures. [35S]-labeled proteoglycans were separated by ion-exchange chromatography followed by gel chromatography on Sephacryl S-500 columns from control (solid line) and cultures exposed to 30 µM CdCl2 (dashed line). The amounts of large (component I) and small (component II) proteoglycans were pooled and calculated as shown by horizontal bars. The recovery was equal to or greater than 85%. Large proteoglycans (component I) were further precipitated with Alcian blue followed by separation by agarose electrophoresis before () and after (+) chondroitinase ABC digestion. PGL = large proteoglycans; PGS = small proteoglycans; GAG = glycosaminoglycan.

View larger version (35K): [in this window] [in a new window]   Figure 3.   Gel chromatography on Sephacryl S-500 and electrophoresis of radiolabeled proteoglycans from the cell layer from control and Cd2+-treated cultures. [35S]-labeled proteoglycans were separated by ion-exchange chromatography followed by gel chromatography on Sephacryl S-500 columns from control (solid line) and cultures exposed to 30 µM CdCl2 (dashed line). The amounts of large (component I), intermediate-sized (component II), and small (component III) proteoglycans were pooled and calculated as shown by horizontal bars. The recovery was equal to or greater than 86%. Large (component I) and intermediate-sized (component II) proteoglycans were further precipitated with Alcian blue followed by separation by agarose electrophoresis before () and after (+) chondroitinase ABC digestion. PGL = large proteoglycans; PGS = small proteoglycans; GAG = glycosaminoglycan.

The large chondroitin/dermatan sulfate proteoglycan recovered from the cell layer similarly accounted for the major proteoglycans eluting close to the void volume, with the remaining fraction consisting of heparan sulfate proteoglycans (Figure 3 inset, component I) likely to be of the perlecan type with a core protein of 350 kD (15). The fraction containing intermediate-sized proteoglycans (Figure 3, fraction II) predominantly comprised small heparan sulfate and some chondroitin/dermatan sulfate proteoglycans, which are likely to be of the biglycan type (Figure 3 inset, component II) (28).

In cells exposed to 30 µM CdCl₂, the most marked effect was observed for the large secreted proteoglycans and for heparan sulfate proteoglycans with a core protein of 250 kD, with synthesis values reduced by up to 67% (Figure 2, component I) compared with media controls. The effect of CdCl₂ was less marked for proteoglycans associated with the cell layer. The production of large cell-associated chondroitin/dermatan sulfate proteoglycans was reduced by 24% (Figure 3, component I), whereas heparan sulfate proteoglycans (most likely of the perlecan type) were reduced by 45%. CdCl₂ also affected the production of intermediate-sized proteoglycans associated with the cell layer, with values reduced by 11 to 24% (Figure 3, component II).

Effect of CdCl₂ on the Production of Small Proteoglycans

Small proteoglycans secreted into the culture media eluted as fraction II (Figure 2) and as fraction III for proteoglycans associated with the cell layer after gel chromatography (Figure 3). In cells exposed to 30 µM CdCl₂, small proteoglycans were reduced by about 48% for both the medium and the cell layer. These small proteoglycans were further separated by hydrophobic interaction chromatography and identified by SDS-PAGE electrophoresis (Figures 4 and 5). Three separate components were identified: an unbound heparan sulfate proteoglycan with a core protein of 35 kD/free glycosaminoglycan chains fraction (17), and two components identified as decorin and biglycan. Decorin was the major small proteoglycan present in the culture medium, whereas the cell layer predominantly contained free glycosaminoglycan (GAG) chains.

View larger version (26K): [in this window] [in a new window]   Figure 4.   Hydrophobic interaction chromatography of radiolabeled small proteoglycans from the culture medium (A) and the cell layer (B) from control and Cd2+-treated cultures. Small proteoglycans were isolated from the medium and the cell layer from control cells (solid line) and cultures exposed to 30 µM CdCl2 (dashed line) by DE-52 and Sephacryl S-500 chromatography; and further separated by hydrophobic interaction chromatography on Octyl Sepharose CL-4B into fractions of heparan sulfate (HSPG)/GAG, decorin, and biglycan. The chromatogram depicts only the incorporation of [35S]. The dotted line represents the guanidinium chloride gradient.

View larger version (71K): [in this window] [in a new window]   Figure 5.   SDS-PAGE of small proteoglycans isolated from the culture medium and the cell layer from control and Cd2+-treated cultures. Small proteoglycans from control () and 30 µM CdCl2 (+)-treated cultures were separated by chromatography on DE-52, Sephacryl S-500, and Octyl Sepharose CL-4B. Proteoglycans were finally subjected to SDS-PAGE electrophoresis. The migration position of the protein standards are indicated on the left. HSPG = heparan sulfate proteoglycan; GAG = glycosaminoglycan; S = start.

The effect of 30 µM CdCl₂ on the production of small proteoglycans associated with the cell layer and secreted into the culture medium was most marked for decorin, with values reduced by up to 65%. CdCl₂ also inhibited the production of other types of proteoglycans isolated by hydrophobic interaction chromatography, with values reduced by 39% for biglycan and by 42% for small heparan sulfate proteoglycans and free GAG chains (Figure 4).

Effect of CdCl₂ on Chondroitin/Dermatan Sulfation Pattern

Table 2 and Figure 6 show results from experiments performed to determine whether these effects of Cd²⁺ on proteoglycan production could be due to undersulfation of chondroitin/dermatan sulfate chains. In these experiments, proteoglycans labeled with [³⁵S]-sulfate and [³H]-glucosamine obtained following purification by ion-exchange chromatography on a DE-52 column were further purified by ion-exchange chromatography on a Mono Q column. The chondroitin/dermatan sulfate chains were subsequently digested by chondroitinase AC/ABC and resulting disaccharides separated into 0-, 4-, and 6-sulfated disaccharides by chromatography on a Lichrosorb NH₂ column (Figure 6). Around 65% of the chondroitin/dermatan sulfate chains were 4-sulfated and there was no difference between control and Cd²⁺-treated cells. However, the percentage of nonsulfated disaccharides was increased from 10.7% for control cells to 20.3% in cells exposed to 30 µM CdCl₂, although the ratio of 4:6 sulfate was unchanged (Table 2, Figure 7) and this decrease in sulfation was much smaller than the observed decrease obtained for proteoglycans with [³⁵S]- labeling. CdCl₂ also affected the specific activity of [³H]- glucosamine incorporated into GAG side chains at the higher dose, because the [³H]:[³⁵S] ratio obtained was increased from 1.6 for media control cells to 3.3 in cells exposed to 30 µM CdCl₂. In contrast, this ratio was unchanged in cells exposed to 10 µM CdCl₂.

View larger version (12K): [in this window] [in a new window]   Figure 6.   Separation of disaccharides from control (A) and Cd2+-treated (B) cultures after chondroitinase AC/ABC digestion of chondroitin/dermatan sulfate on a Lichrosorb NH2 column. Proteoglycans were separated by ion-exchange chromatography on a DE-52 column. Any remaining hyaluronan was removed by ion-exchange chromatography on a Mono Q column. The sulfation patterns in the GAG chains were further determined after digestion with chondroitinase AC/ABC, followed by separation of the degradation products on a Lichrosorb-NH2 column as described in MATERIALS AND METHODS. Both the incorporation of [35S] (dashed line) and [3H]-glucosamine (solid line) are shown.

View larger version (37K): [in this window] [in a new window]   Figure 7.   Effect of Cd2+ on fibroblast 1(I) procollagen and proteoglycan core protein gene expression. Total RNA was isolated and separated by electrophoresis on 1% agarose gel (approximately 10 µg RNA/well). The expression of the housekeeping protein GAPDH is also shown. The size of the transcripts are shown on the right.

Effect of CdCl₂ on Procollagen and Proteoglycan Core Protein Steady-State mRNA Levels

Figure 7 shows the effect of Cd²⁺ on mRNA steady-state levels for 1(I) procollagen and several proteoglycan core proteins. At the lower dose, CdCl₂ had no effect on fibroblast procollagen mRNA steady-state levels, whereas these levels were reduced by around 37% for cells exposed to 30 µM CdCl₂. Cd²⁺ exposure also led to a reduction in decorin and versican mRNA steady-state levels with levels reduced by 40 and 57%, and by 61 and 81% for cells exposed to 10 µM and 30 µM CdCl₂, respectively. In contrast, biglycan mRNA steady-state levels were totally unaffected at 10 µM CdCl₂ and only slightly reduced at 30 µM CdCl₂.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of Cd²⁺ Fibroblast Proteoglycan Synthesis

In this paper we show that Cd²⁺, at noncytotoxic doses, is a potent inhibitor of fibroblast proteoglycan production in vitro, with values reduced by about 40% in cells exposed to 30 µM CdCl₂. In contrast, total protein synthesis was unaffected at both doses of Cd²⁺ examined, indicating that the inhibition obtained was selective. For comparision, procollagen production by these cells was also decreased at both doses of Cd²⁺ examined, with maximal inhibitory effects of around 70% at 30 µM CdCl₂, indicating that procollagen synthesis is more sensitive to the inhibitory effects of Cd²⁺ than proteoglycan synthesis. However, detailed analysis of the different proteoglycans affected by Cd²⁺ revealed that the magnitude of inhibition was most pronounced for the major matrix-associated proteoglycans, versican and decorin, as well as the large heparan sulfate proteoglycans, with synthesis values reduced by between 60 and 70%.

Effect of Cd²⁺ on Proteoglycan Sulfation

Sulfation experiments performed showed that Cd²⁺ altered the specific activity of the UDP-N-acetylhexosamine pool, which manifests itself as an increase in the ratio of [³H]- glucosamine to [³⁵S]-sulfate in unsaturated dissaccharides. The mechanism by which Cd²⁺ induces these changes in UDP-N-acetylhexosamine metabolism is not clear, but Cd²⁺ has been reported to have dramatic effects on the metabolism of glycolytic intermediates (29). In contrast, Cd²⁺ had no effect on the specific activity of the sulfate pool. This may be due to the fact that the only major influx of sulfate into the intracellular sulfate pool is from the culture medium. This exchange is rapid and a constant specific activity is reached within minutes, both in the sulfate and the 3'-phosphoadenosine-5'phosphosulfate pool (30). [³⁵S]-sulfate-labeling therefore appears to be a more reliable method for measuring GAG production than does [³H]-glucosamine labeling.

Our results are not in agreement with those obtained by Kaji and colleagues (31), who showed that Cd²⁺ increases GAG synthesis by various cell types, including fibroblasts, by increasing the incorporation of [³H]-glucosamine into GAG chains. However, their experiments were performed at much lower concentrations of Cd²⁺ (2 µM) than used here, and any potential changes in the specific activity of the UDP-N-acetylhexosamine pool were not taken into consideration. However, in accord with our results, these researchers similarly showed that Cd²⁺ decreases the incorporation of [³⁵S]-sulfate into GAG chains, especially for GAGs of the heparan sulfate type. This was interpreted as indicating that Cd²⁺ was inhibiting sulfation. However, after performing detailed analysis, our experiments showed that GAG chains were only slightly undersulfated (from 0 to 20%), so that the inhibitory effects of Cd²⁺ on proteoglycan synthesis, based on the incorporation of [³⁵S]-sulfate reported here, are unlikely to be due to undersulfation alone.

Effect of Cd²⁺ on Fibroblast Procollagen and Proteoglycan mRNA Steady-State Levels

The data obtained by Northern blot analysis indicate that the effects of Cd²⁺ at the protein level were occurring, at least in part, via a reduction in mRNA steady-state levels. However, in certain instances the magnitude of inhibition obtained at the protein level was greater than that obtained at the mRNA level, indicating that Cd²⁺ may also be acting via translational or post-translational mechanisms.

In Vivo Relevance of These Findings

This is, to our knowledge, the first detailed analysis of the effects of Cd²⁺ on fibroblast proteoglycan metabolism and gene expression, and the results obtained are consistent with our hypothesis that Cd²⁺ may compromise several key processes of connective tissue repair. Proteoglycans, as well as influencing tissue compliance and fluid balance, perform a number of other regulatory functions by virtue of their involvement in such diverse functions as the regulation of cell proliferation, migration, and adhesion. The effects of Cd²⁺ on versican and decorin are of particular interest. Versican is usually localized between collagen fibrils and interacts with hyaluronan to form large extracellular aggregates. Decorin is thought to play a pivotal role in the organization of newly formed connective tissue by influencing collagen fibrillogenesis and architecture (32), and has also been proposed as a modulator of transforming growth factor- action (33). In addition, specific decreases in decorin production and gene expression have also been reported during the development of other lung disorders associated with perturbations in connective tissue metabolism, including bleomycin-induced pulmonary fibrosis (19). The experiments reported here suggest that Cd²⁺ may not only prevent the deposition of connective tissue proteins following Cd²⁺ exposure, but may also play a role in influencing the abnormal spatial organization of newly synthesized connective tissue components, commonly observed in emphysema (34). Our findings may furthermore also provide an explanation for the in vivo results obtained by Radhakrishnamurthy and colleagues (35) showing that proteoglycan levels were decreased in emphysema induced following administration of CdCl₂ in experimental animals.

The results reported in this paper may be particularly relevant to cigarette smokers, who represent the largest group of individuals who go on to develop emphysema. Smoking represents a major source of repeated low-dose exposure to Cd²⁺ with an estimated biologic half-life of 10 yr (36), and there are striking similarities between the lesions observed in experimental animals chronically exposed to Cd²⁺ and centrilobular emphysema in cigarette smokers (4). In terms of the in vivo relevance of the doses of Cd²⁺ used in our experiments, we estimated from previous reports, where Cd²⁺ was measured as a concentration per dry weight (36), that the concentrations of Cd²⁺ in the lungs of patients with moderate to severe emphysema would be in the micromolar range. We would predict that local concentrations at sites particularly vulnerable for deposition may even be higher.

Conclusions

In this study we have shown that Cd²⁺, at low noncytotoxic doses likely to be attained and sustained in the lung, selectively inhibits procollagen and proteoglycan synthesis and gene expression and alters the sulfation of proteoglycans produced. We propose that these effects of Cd²⁺, in the face of ongoing oxidant and proteinase-mediated connective tissue damage, may tip the balance between connective tissue protein synthesis and degradation in favor of degradation and may therefore play an important role in the pathogenesis of emphysema in humans exposed to Cd²⁺ fumes occupationally or in cigarette smoke.